Abstract

Sensitivity and ambiguity of rock properties to different reservoir conditions is analyzed for clastic reservoirs through well-log modeling. The analysis shows that; for the simple case of two lithologies (sand and shale), two fluids (gas and brine) and changes in porosity; the use of two rock properties, as routinely used in cross-plotting techniques, is ambiguous; that is: more than one set of petrophysical properties result in the same rock properties. Incorporating a third independent rock property, although reduces ambiguity, does not guaranty that a set of independent rock properties (i.e. P-wave velocity (VP), S-wave velocity (VS) and density (?)) will result in a unique set of petrophysical properties (i.e. porosity (?), shale volume (VSH) and water saturation (SW)). Sensitivity analysis shows that changes in water saturation result in the least change in rock properties as compared to changes in rock properties when porosity or lithology are modified; additionally small changes in lithology and porosity cancelout the effect that a large change in water saturation has on rock properties.

Introduction

Seismic attributes have the objective, amongst others, of reducing the risk when determining petrophysical properties using surface measurements. The degree of success in this task is related to the number of independent attributes computed from the seismic data as well as the data used in their calibration. Two approaches in risk reduction are commonly undertaken: one in which a large amount of attributes are computed and then statistically related to well properties, and the other in which rock properties directly related to the rocks' petrophysical attributes are estimated and then used to deterministically estimate the petrophysical properties of interest (i.e. porosity, lithology and pore fluids). Whichever approach is taken, the fact remains that seismic data is a function of changes of three rock properties (P- and S-wave velocity and density (VP, VS and ? respectively)) across interfaces and the usefulness of attributes in reducing risk is dependent upon the ambiguity and sensitivity of these physical properties to the petrophysical properties of interest.

Two attributes related to petrophysical properties are commonly obtained through AVO (Shuey's1 Ro and Go, Rp and Rs, etc.) and their analysis is routinely done by crossplotting these attributes or rock properties derived from these through inversion (P- and S-wave impedance or LambdaRho and MuRho for example). In these cross-plots, anomalous points (away from a "background trend") are identified and, depending upon the out-layer position, associated to fluid changes within reservoirs. The technique assumes that only fluid type and/or saturation change within the reservoir and that other petrophysical properties (i.e. lithology and porosity) remain constant or do not modify rock properties. The following analysis explores the possibility of having different petrophysical properties produce the same rock properties (ambiguity); as well as how much deviation there is from the "background trend" when different petrophysical properties are modified. Sensitivity of P- and S-wave impedances are analyzed in the understanding that other properties (P- and Swave reflectivity, LambdaRho and MuRho, etc.) can be derived from these.

This content is only available via PDF.
You can access this article if you purchase or spend a download.